The Journal of the Korea institute of electronic communication sciences (한국전자통신학회논문지)

Korea Institute of Electronic Communication Science (한국전자통신학회)
권호 표지
DOI QR코드

DOI QR Code

Comparison Speed of Pedestrian Detection with Parallel Processing Graphic Processor and General Purpose Processor

병렬처리 그래픽 프로세서와 범용 프로세서에서의 보행자 검출 처리 속도 비교

  • Received : 2014.11.12
  • Accepted : 2015.02.09
  • Published : 2015.02.28
Download PDF
⟨ Previous Next ⟩

Abstract

Video based object detection is basic technology of implementing smart CCTV system. Various features and algorithms are developed to detect object, however computations of them increase with the performance. In this paper, performances of object detection algorithms with GPU and CPU are compared. Adaboost and SVM algorithm which are widely used to detect pedestrian detection are implemented with CPU and GPU, and speeds of detection processing are compared for the same video. As results of frame rate comparison of Adaboost and SVM algorithm, it is shown that the frame rate with GPU is faster than CPU.

영상기반 객체 검출은 지능형 CCTV 시스템을 구현하는데 있어 기본적인 기술이다. 객체 검출을 위하여 다양한 특징점과 알고리즘이 개발되었으나, 성능에 비례하여 계산량이 많다. 본 논문에서는 GPU와 CPU를 활용하여 객체 검출 알고리즘의 성능을 비교하였다. 일반적으로 보행자 검출에 널리 사용되고 있는 Adaboost 알고리즘과 SVM 알고리즘을 각각 CPU와 GPU에 맞도록 구현하고 동일 영상에 대하여 검출 처리 속도를 비교하였다. Adaboost 알고리즘과 SVM 알고리즘에 대하여 처리 속도를 비교한 결과 GPU가 CPU에 비하여 약 4 배 정도 빠른 처리를 할 수 있음을 확인하였다.

Keywords

References

  1. H-M. Moon and S-B. Pan, "The human identification method in video surveillance system," J. of The Korean Institute of Information Technology, vol. 8, no. 5, May 2010, pp. 199-206.
  2. H-M. Moon and S-B. Pan, "The analysis of de-identification for privacy protection in intelligent video surveillance system," J. of The Korean Institute of Information Technology, vol. 9, no. 7, July 2011, pp. 189-200.
  3. H.-T. Kim, G.-H. Lee, J.-S. Park, and Y.-S. Yu, "Vehicle detection in tunnel using Gaussian mixture model and mathematical morphological processing," J. of The Korean Institute of Electronic Communication Sciences, vol. 7, no. 5, Oct. 2012, pp. 967-974.
  4. M.-W. Kim, C.-M. Oh, D. Aurrahman, Y.-G. Ahn, and C.-W. Lee, "The virture screen using skin tone and GMM foreground segmentation," In Proc. Conf. of The Korea Information Processing Society, vol. 15, no. 1, May 2008, pp. 179-181.
  5. C. Stauffer and W. E. L. Grimson, "Adaptive background mixture models for real-time tracking," In Proc. IEEE Computer Vision and Pattern Recognition(CVPR) 1999, vol. 2, June 1999, pp. 246-252.
  6. A. Elgammal, D. Harwood, and L. S. Davis, "Non-parametric model for background subtraction," In Proc. European Conf. on Computer Vision(ECCV 2000), vol. 1843, June. 2000, pp. 751-767.
  7. T. Ahonen, A. Hadid and M. Pietikaninen, "Face description with local binary patterns : application to face recognition," IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 28, no. 12, Dec. 2006. pp. 2037-2041. https://doi.org/10.1109/TPAMI.2006.244
  8. D. Geronimo, A. D. Sappa, A. Lopez and D. Ponsa, "Pedestrian detection using Adaboost learning of features and vehicle pitch estimation," In Proc. Int. Conf. Visualization, Imaging and Image Processing(IASTED), Palma de Mallorca, Spain, Aug. 2006. pp. 400-405.
  9. Thorsten Joachims, "Training linear SVMs in linear time," In Proc. of Int. conf. on Knowledge Discovery and Data Mining(KDD), Philadelphia Pennsylvania, Aug. 2006. pp. 217-226.
  10. "NVIDIA Tegra K1, A New Era in Mobile Computing," White-paper of NVIDIA, 2013.